[spankbang] Fix and improve metadata extraction

This commit is contained in:
Sergey M․ 2019-07-14 00:21:39 +07:00
parent 5f562bd4bb
commit f9eeeda31c
No known key found for this signature in database
GPG Key ID: 2C393E0F18A9236D
1 changed files with 17 additions and 10 deletions

View File

@ -5,6 +5,7 @@ import re
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ( from ..utils import (
ExtractorError, ExtractorError,
merge_dicts,
orderedSet, orderedSet,
parse_duration, parse_duration,
parse_resolution, parse_resolution,
@ -26,6 +27,8 @@ class SpankBangIE(InfoExtractor):
'description': 'dillion harper masturbates on a bed', 'description': 'dillion harper masturbates on a bed',
'thumbnail': r're:^https?://.*\.jpg$', 'thumbnail': r're:^https?://.*\.jpg$',
'uploader': 'silly2587', 'uploader': 'silly2587',
'timestamp': 1422571989,
'upload_date': '20150129',
'age_limit': 18, 'age_limit': 18,
} }
}, { }, {
@ -113,26 +116,29 @@ class SpankBangIE(InfoExtractor):
self._sort_formats(formats) self._sort_formats(formats)
info = self._search_json_ld(webpage, video_id, default={})
title = self._html_search_regex( title = self._html_search_regex(
r'(?s)<h1[^>]*>(.+?)</h1>', webpage, 'title') r'(?s)<h1[^>]*>(.+?)</h1>', webpage, 'title', default=None)
description = self._search_regex( description = self._search_regex(
r'<div[^>]+\bclass=["\']bottom[^>]+>\s*<p>[^<]*</p>\s*<p>([^<]+)', r'<div[^>]+\bclass=["\']bottom[^>]+>\s*<p>[^<]*</p>\s*<p>([^<]+)',
webpage, 'description', fatal=False) webpage, 'description', default=None)
thumbnail = self._og_search_thumbnail(webpage) thumbnail = self._og_search_thumbnail(webpage, default=None)
uploader = self._search_regex( uploader = self._html_search_regex(
r'class="user"[^>]*><img[^>]+>([^<]+)', (r'(?s)<li[^>]+class=["\']profile[^>]+>(.+?)</a>',
r'class="user"[^>]*><img[^>]+>([^<]+)'),
webpage, 'uploader', default=None) webpage, 'uploader', default=None)
duration = parse_duration(self._search_regex( duration = parse_duration(self._search_regex(
r'<div[^>]+\bclass=["\']right_side[^>]+>\s*<span>([^<]+)', r'<div[^>]+\bclass=["\']right_side[^>]+>\s*<span>([^<]+)',
webpage, 'duration', fatal=False)) webpage, 'duration', default=None))
view_count = str_to_int(self._search_regex( view_count = str_to_int(self._search_regex(
r'([\d,.]+)\s+plays', webpage, 'view count', fatal=False)) r'([\d,.]+)\s+plays', webpage, 'view count', default=None))
age_limit = self._rta_search(webpage) age_limit = self._rta_search(webpage)
return { return merge_dicts({
'id': video_id, 'id': video_id,
'title': title, 'title': title or video_id,
'description': description, 'description': description,
'thumbnail': thumbnail, 'thumbnail': thumbnail,
'uploader': uploader, 'uploader': uploader,
@ -140,7 +146,8 @@ class SpankBangIE(InfoExtractor):
'view_count': view_count, 'view_count': view_count,
'formats': formats, 'formats': formats,
'age_limit': age_limit, 'age_limit': age_limit,
} }, info
)
class SpankBangPlaylistIE(InfoExtractor): class SpankBangPlaylistIE(InfoExtractor):